Andreas Lee

Three dimensional ultrasound: abnormalities of the fetal face in surface and volume rendering mode

Objective To evaluate three dimensional ultrasound in surface and volume rendering mode in fetal malformations of the face.

Neural networks for recognizing patterns in cardiotocograms.

The cardiotocogram (CTG) is commonly used for routine fetal monitoring in the delivery room. A major problem is that the interpretation of the CTG trace requires experienced specialists. In order to avoid long gaps between the detection of a suspicious pattern and the intervention, the CTG has to be checked in short intervals. An automated monitoring system at the obstetric site can reduce such delays. Therefore, an alarm system immediately reporting suspicious events has been built. The focus of our study was put on the question whether AI techniques such as neural networks are suited to the task of recognizing patterns in the CTG trace. In a comparative study, their performance was evaluated against that of conventional methods. The neural networks turned out to provide significantly better results than the tested conventional methods.

Endometrial volume change during spontaneous menstrual cycles : volumetry by transvaginal three-dimensional ultrasound

OBJECTIVE At present, only limited data are available on endometrial volume during the menstrual cycle. Most of these studies deal with animal models and use magnetic resonance imaging for volume measuring. The application of three-dimensional ultrasound in endometrial volume estimation is the subject of this study. SETTING Patients visiting the outpatient unit of the division of endocrinology and reproductive medicine of a university hospital. PATIENT(S) Twenty patients with a history of a normal menstrual cycle were selected. INTERVENTION(S) Ultrasound examinations were performed during a single menstrual cycle in addition to routine laboratory tests. MAIN OUTCOME MEASURE(S) Uterus-endometrial volume ratio. RESULT(S) Data from 18 patients could be evaluated. In 81 examinations the endometrium volume could be determined. Mean endometrial volume measured by three-dimensional ultrasound was 1.23 cm3. Mean uterus volume was 48.93 cm3. The change of the uterus-endometrial volume ratio showed a good correlation with the day of menstrual cycle. Quadratic regression analysis of volume and cycle length was R2 = 0.432. CONCLUSION(S) Three-dimensional ultrasound allows assessment of volume data of the female internal genitalia. In this study changes of the endometrial volume in menstrual cycles were measured. Additional studies are required to give information on the clinical impact of this new technique of endometrial volume estimation.

Three-dimensional ultrasonographic imaging of fetal tooth buds for characterization of facial clefts

The purpose of this prospective study was to investigate whether the antenatal characterization of fetal facial clefts can be improved by three-dimensional ultrasonographic visualization of fetal tooth buds. Between January 1996 and June 1998, seventeen consecutive fetuses with facial clefts were examined for fetal maxillary tooth buds in the cleft area using three-dimensional multiplanar reconstruction. It was possible in all cases to classify the clefts either as cleft lip alone or unilateral cleft lip and palate or bilateral cleft lip and palate. Three-dimensional computed tomography and histological jaw sections of three stillborn infants were produced in order to examine the correlation between the sonographic, radiographical and histological findings. The prenatal characterization of the facial clefts by means of a visualization of the tooth buds showed to be accurate postnatally in all cases. The sonographic proof of tooth buds might gain increasing importance as this technique seems to facilitate and improve the prenatal classification of suspected facial clefts.

Application of artificial neural networks for detection of abnormal fetal heart rate pattern: a comparison with conventional algorithms

Cardiotocography signals were sampled during labour in 53 patients. A recurrent artificial neural network with hidden layer feedback was trained and performance was compared with that of several conventional systems. Correct and false positive rates of all systems tested were calculated. To ensure that the performance of neural networks was not just caused by using different cut-off levels, the threshold used for conventional methods were also adapted and optimised. The correct positives rate of neural networks was between 0.72 and 0.9, and the false positive rate between 0.2 and 0.4. Before optimising, conventional algorithms produced a very low correct positive (0.02-0.5) and a low false positive rate (0.0-0.08). After adjusting the parameters, the tested neural networks still performed better than optimised conventional systems.